In a cellular communications network, user equipment (UE) (such as mobile telephones, mobile devices, mobile terminals, etc.) can communicate with other user equipment and/or remote servers via base stations. LTE systems include an Evolved Universal Terrestrial Radio Access Network (E-UTRAN) and an Evolved Packet Core (EPC) network (or simply ‘core network’). The E-UTRAN includes a number of base stations (‘eNBs’) for providing both user-plane (e.g. Packet Data Convergence Protocol (PDCP), Radio Link Control (RLC), Medium Access Control (MAC) and PHYsical (PHY) layers) and control-plane (e.g. Radio Resource Control (RRC)) protocol terminations towards the UE.
Recent developments in communication networks have seen increased deployment of so called ‘small’ cells operated by Low Power Nodes (LPNs), such as pico eNBs, femto eNBs, home eNBs (HeNBs) or the like, which cells have a smaller coverage area than existing macro cells operated by a higher power (regular) macro base station. Networks comprising a number of different cell types, for example a network comprising a macro cell and a femto cell, are referred to as Heterogeneous Networks, or HetNets. In the following description the term base station is used to refer to any such macro base station or LPN. Conventionally, a mobile telephone is configured to communicate via one base station (using an associated radio link). However, in a study on small cell enhancements for E-UTRA and E-UTRAN (3GPP technical report (TR) no. 36.842, the contents of which are incorporated herein by reference), a so-called ‘dual connectivity’ functionality was introduced to improve, for example, the coverage of high data rates for user equipment, temporary network deployment, cell edge throughput and/or to increase system throughput. The dual connectivity feature established techniques for compatible mobile telephones (and other user equipment) to communicate with multiple network points, substantially simultaneously. Specifically, this ‘dual connectivity’ functionality refers to an operation mode where a given mobile telephone (operating in RRC_CONNECTED mode) consumes radio resources provided by at least two different network points (e.g. two or more base stations). Typically, one of the network points involved in the dual connectivity functionality is a macro base station and the other network point (or a plurality of network points) comprises a low power node (or plurality of low power nodes).
Each network point (also referred to as ‘access point’) involved in the provision of dual connectivity for a mobile telephone may assume a different role. One of the network points may be referred to as a master base station (MeNB) and each one of the other network points may be referred to as a secondary base station (SeNB). Typically, the various secondary base stations involved in the provision of dual connectivity are coupled (to the MeNB and hence the core network) via a so-called non-ideal backhaul. Further, in a dual connectivity scenario, one of the base stations (the MeNB) routes control plane signalling to the core network via an associated interface (e.g. the S1 interface), regardless of whether or not the other base station is also connected to the core network for user plane communication (e.g. to a serving gateway).
The MeNB/SeNB roles do not necessarily depend on each base station's capabilities/type (e.g. power class) and may be different for different mobile telephones (even when using the same base stations).
In accordance with the dual connectivity functionality, a mapping between the mobile telephone's radio (communication) bearer(s) and the base stations may be realised as follows:                a so-called Master Cell Group (MCG) bearer in which a radio bearer is served by the MeNB only (or ‘MeNB-specific bearer’);        a so-called Secondary Cell Group (SCG) bearer in which a radio bearer is served by the SeNB only (or ‘SeNB-specific bearer’); and        a split bearer in which a radio bearer is served by the MeNB and the SeNB.        
In order to ensure that an appropriate level of service (e.g. a desired data rate) can be provided for each user in the communication network, the network operator assigns various parameters that determine an aggregate maximum bit rate (AMBR) that can be provided to the users (subscribers) in the network per subscriber and per access point. Specifically, for each subscriber, the Home Subscriber Server (HSS) holds an associated ‘HSS_APN-AMBR’ parameter (per APN) and an ‘HSS_UE-AMBR’ parameter, forming part of the user's subscription data.
The HSS_APN-AMBR (APN Aggregate Maximum Bit Rate) parameter for a particular (subscriber's) user communication device limits the non-guaranteed aggregate bit rate across all PDN connections by that user communication device via a particular APN. The actual ‘APN-AMBR’ parameter to be used (enforced) by the given access point (e.g. P-GW) is provided by the MME based on subscription data obtained from the HSS.
The HSS_UE-AMBR (UE Aggregate Maximum Bit Rate) parameter for a particular (subscriber's) user communication device limits the total traffic of that user communication device on uplink and downlink (via the serving base station). The actual ‘UE-AMBR’ parameter to be used (enforced) by the serving base station is provided by the MME based on subscription data obtained from the HSS. Specifically, the MME computes the UE-AMBR parameter such that it equals the smaller of the sum of all HSS_APN-AMBR parameters of active APNs and the HSS-UE-AMBR parameter. This is further illustrated in the 3GPP TS 23.401 standard, the contents of which are incorporated herein by reference. The MME transmits the calculated UE-AMBR parameter to the serving base station, which base station is thus able to allow/discard data traffic for the user communication device in accordance with the UE-AMBR parameter. This is further illustrated in the 3GPP TS 36.413 and TS 36.300 standards, the contents of which are incorporated herein by reference.
Thus traffic sent/received by a particular user communication device in excess of the bit rate indicated by the UE-AMBR parameter may get discarded by a rate shaping function of the base station serving that user communication device, and traffic exceeding the bit rate indicated by the applicable APN-AMBR parameter may get discarded by a rate shaping function of the corresponding APN. The UE-AMBR parameter and the APN-AMBR parameter are applicable across all non-Guaranteed Bit Rate (non-GBR) bearers of a particular subscriber (i.e. a user communication device associated with that subscriber).
Each base station guarantees a downlink guaranteed bit rate associated with a so-called guaranteed bit rate (GBR) bearer, enforces a downlink maximum bit rate (MBR) associated with a particular GBR bearer and enforces a downlink Aggregate Maximum Bit Rate (AMBR) associated with a group of non-GBR bearers. Further, in the uplink, by limiting the total grant of communication resources to an item of user equipment, the base station can ensure that a UE-AMBR for a respective group of non-GBR bearers associated with each item of user equipment, plus the sum of MBRs is not exceeded.
There is a general consensus that, during dual connectivity, the MeNB should manage the UE-AMBR and provide, to the SeNB, information which assists the SeNB to provide both downlink and uplink AMBR enforcement when the SCG bearer option is applied.
The overall UE-AMBR enforced for a particular dual connectivity UE may be split between an MeNB-specific UE-AMBR (MUE-AMBR) for that UE and an SeNB-specific UE-AMBR (SUE-AMBR) for that UE. The SUE-AMBR is sent to the SeNB by the MeNB managing the overall UE-AMBR, and the SeNB enforces the SUE-AMBR accordingly.
However, the inventors have realised that, as a result of non-GBR bearers of a UE potentially being distributed between the MeNB and SeNB during dual connectivity, and contrary to the current consensus, the generally accepted route forward can result in a sub-optimal solution in which, for example, the UE-AMBR is not always split between the master and secondary base stations in the most efficient manner.
For example, when there is a lot of data in SeNB's buffer which is not sent due to enforcement of the SUE-AMBR, and the data rate of data arriving in MeNB is significantly lower than MUE-AMBR, then the actual data rate available to the UE may be significantly lower than the overall UE-AMBR that the UE is entitled to by contract. Similarly, when there is a lot of data in MeNB's buffer which is not sent due to enforcement of the MUE-AMBR, and the data rate of data arriving in SeNB is significantly lower than SUE-AMBR, then the actual data rate available to the UE may be significantly lower than the overall UE-AMBR that the UE is entitled to by contract. Thus, the UE may suffer data loss due to UE-AMBR enforcement unnecessarily.
To help illustrate this issue, a number of examples of unnecessary data loss are summarises below:
Downlink—Unnecessary Packet Dropping in SeNB:
For a particular UE, the value of UE-AMBR may be e.g. 10 Mbps for distribution between the two radio bearers used by the UE (e.g. E-RAB#1 provided via the MeNB and E-RAB#2 provided via the SeNB). In this case, for example, the following parameters may be configured (for the UE's non-GBR bearers):
MUE-AMBR=5 Mbps (for communication bearers over E-RAB#1 via the MeNB)
SuE-AMBR=5 Mbps (for communication bearers over E-RAB#2 via the SeNB)
However, when E-RAB#1 has almost no activity but the communication bearers over E-RAB#2 carry a large amount of data, it is possible that the aggregated data rate for E-RAB#2 may exceed the allowed 5 Mbps. In this case, therefore, the SeNB enforces the SUE-AMBR by dropping data packets for the UE that are determined to be above the user's allowance (SUE-AMBR). This may result, from a user point of view, in the UE only receiving an effective 5 Mbps (assuming that E-RAB#1 has no activity and SUE-AMBR for E-RAB#2 is set to 5 Mbps) while the contracted UE-AMBR is 10 Mbps.Downlink—Unnecessary Packet Dropping in MeNB:
Similarly, using the parameters of the previous example (MUE-AMBR=5 Mbps and SUE-AMBR=5 Mbps), there may be scenarios in which the MeNB may drop data packets for a particular UE unnecessarily.
For example, when E-RAB#2 via the SeNB has almost no activity but the communication bearers provided over E-RAB#1 carry a large amount of data, it is possible that the aggregated data rate for E-RAB#1 may exceed the allowed 5 Mbps via that base station (MeNB). The MeNB may thus start dropping data packets resulting, from a user point of view, in the UE only receiving an effective 5 Mbps while its contracted UE-AMBR is 10 Mbps.
It will be appreciated that similar scenarios are also possible for the enforcement of uplink UE-AMBRs.
In summary, when the communication bearers for a particular UE in dual connectivity exhibit an imbalance (at least temporarily) between the base stations serving the UE, it may be difficult or impossible to ensure that the aggregated data rate for a particular UE meets the data rate (UE-AMBR) associated with the user's subscription.